Is it not the case that when the coefficients have a t-distribution that the ols estimates have minimum variance and thus maximise the t-statistics. If you are trying to fit a heavy tailed distribution to some phenomenum and wis to estimate the appropriate degrees of freedom be sure to use a form of generalised t-distribUtion as there are constraints on the mean and variance of the standard t-distribution.
On Wednesday, 21 December 2011, clarodina@lycos.com <clarodina@lycos.com> wrote:
>
> Hi PS,
>
> Want to have the coefficients maximising t value
>
> Matlab have a add on doing the maximisation
>
> How to do the maximising on gretl?
>
> Clarodina
> _______________________________________________
> Gretl-users mailing list
> Gretl-users@lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>
--
John C Frain
Economics Department
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.html
mailto:frainj@tcd.ie
mailto:frainj@gmail.com